NOHARM study finds medical AI often matches or beats general practitioners
jonc101x · x · 2026-07-27
In the NOHARM study, frontier and medical-specific AI systems handled complex clinical consultations remarkably well, often matching or outperforming general practitioners — even when the doctors had references or AI assistance themselves.
The thread argues that the real risk is not just confidently wrong answers, but omission: reassuring users with "nothing to do" when a dangerous condition actually needs a workup.
Related event: AI Health Advice Is Already Here; Safety Systems Are the Real Bottleneck(16 posts)→
More from Safety
- Meta Muse's first suggested name matches user's childhood dog, raising privacy questions — matt_slotnick · 2026-09-23
- Open-source advocates call doom narratives a regulatory moat against open weights — AlexTensor · 2026-09-23
- AI safety will follow engineering tradition: formal proofs for simple cases, evals for complex — burny_tech · 2026-09-23
- Stochastic Parrots authors rebut AI-pause letter: focus on present harms, not sci-fi risk — marigo · 2026-09-23
- Devs mock labs' cyber-enabled Claude/GPT testing as 'felonies sold as safety research' — ctjlewis · 2026-09-23
- Okta launches Human Principal, binding AI agents to verified humans via World ID — BecauseCulture · 2026-09-23